Parts 1 and 2 are here and here.
5. Corruption (22:40). Bennet asks how sortition prevents the influence of money. Troesoyer says that sortition makes it easier to prevent corruption. In his argument he refers to two factors: term length and avoiding the electoral need to raise campaign money. Term length is important, Troesoyer says, because decision makers working together for a long period of time are able to develop trust between themselves. This allows them to collaborate among themselves and with other long-term actors, and thus to collude on corrupt enterprises. Campaign finance is a fertile ground which fosters corrupt collusion. Actors can promise help in terms of campaign funds in exchange for promotion of their interests. They can also make threats to support competitors in retaliation for non-cooperation.
But neither of these factors make for a convincing advantage of sortition over elections. Electoral systems can be, and often are, set up so that campaigns are highly regulated and publicly financed, greatly reducing the ability of campaign funds to inflence outcomes. The fundamental reason that elected officials serve narrow interests is not a direct quid-pro-quo but rather that the elected officials have a world-view in which the narrow interests are legitimate and should be served. Granted, campaign funds can be a corrupting factor but that can be fixed. The problem is that fixing that does not fix the fundamental problem with elections.
As for term lengths, the implied assumption in the argument is that allotted bodies have “short terms”. But short terms imply not much time to develop familiarity with issues, which is hardly a tool for better decision making. Furthermore, specifically regarding trust, in any setup where members of a group are expected to communicate with each other and with others (such as advisors and experts), trust is an asset, not a problem. Such trust should not be thought of as allowing corrupt collusion but allowing fruitful collaboration. For this reason allotted bodies, and certainly any allotted body that makes proposals (rather than takes up-or-down votes on pre-made proposals), must have terms measured in years. Thus, unless “short terms” is measured in years, allotted bodies (or at least some of them) should not have short terms of office.
It is true that elected candidates and officials spend entire lifetimes developing connections within the system and within the ruling class. It is true also that these lifetimes spent in these circumstance do play an important role in shaping that world-view referred to above as the foundation of corruption. Thus, to some extent it is true that the time spent inside decision making circles is crucial. But the difference is between a period of years and a period spanning a lifetime rather than between a period of days or weeks and a period of years.
6. A proposers-deciders division of labor (27:30). Troesoyer proposes Terry Bouricius’s plan for having two bodies involved in legislation. The first being a longer-term body (serving for weeks or months) sampled from volunteers which comes up with a bill. The second being a “quasi-mandatory” short-term body (working for days) making an accept or reject decision on the proposed bill based on input from the proposers.
This is all round a truly bad idea. Having the proposals made by “volunteers” makes it very likely they would be highly influenced by narrow interests. Having the decisions made by short-term bodies makes it very likely those decisions would be less than well-informed well-considered since becoming informed and weighing considerations takes time. It is likely that rhetorical manipulation would play an important role in decisions that require understanding complex issues but are made within a very limited time. Furthermore, forcing people into decision-making bodies may make those bodies appear representative, but in no way makes it likely that the bodies would generate decisions that really do represent the informed and considered will of the population. Those people who do not really wish to be present would likely not invest much effort in making their decision. At best they would abstain. More dangerously they may very well be targets for manipulative persuasion, and might even be deliberately non-cooperative.
The idea of having a proposers-deciders division of labor is by itself rather questionable. Bouricius’s claim that people’s handiwork is better judged by others than by themselves seems rather anti-democratic since it implies that people cannot represent their own interests as well as others can. But be the answer to this question what it may, the Bouricius proposal goes beyond a division of labor. The idea that such a division of labor would allow transmuting two non-democratic bodies (one which is not representative and the other which does not have the time to properly undertstand the matters on which it decides) into a single democratic process is pure political alchemy.
Adam Cronkright
Ali
Ahmed R. Teleb
avanderven
avergne
Alex Kovner
azakaras
Brett Hennig
campbellwallace
Simon Threlkeld
Nick Coccoma
codyhipskind
Common Lot Sortitionist
Conall Boyle
constitutionalism
davidschecter
Ethan McCutchen
graytmatter
gtridimas
Henry Tuttle
H.J. Hofkirchner
HpR
James Kierstead
John Burnheim
jgarry20151
Jochen Krattenmacher
johnedwardtaylor
johnhuang865
Jonathan Crock
keithsutherland
Jacob Richter
kkoreilly
lukewhito
SimonPek
Nathan El
Nicholas Gruen
Owen Shaffer
Patrick Chalmers
Paul Gölz
peterstone
Pierre-Etienne Vandamme
pronfeld
Emil
rcaze
richwbrown
Roger Knights
roslynfuller
Pierre-Etienne Vandamme
shaharavin
tbouricius
Ted Wachtel
Joshua Laferriere
theodorestathis
towardsautonomyblog
Wayne Liebman
Yoram Gat
Oliver Milne
Yoram>: The fundamental reason that elected officials serve narrow interests is not a direct quid-pro-quo but rather that the elected officials have a world-view in which the narrow interests are legitimate and should be served.
That is certainly the foundational principle of Gatism (that you share with a tiny number of unreconstructed Marxists), but it would help to hear some supportive argumentation. It’s perfectly possible for someone to offer themselves for election in order to serve the interests of their constituents.
LikeLiked by 1 person
A couple of counter-examples: do you deny that when Keir Hardy founded the Labour Representation Committee in 1900, he (and his colleagues) were not genuinely seeking to represent the interests of the industrial working classes? Or that the Haileybury and Oxford-educated former barrister and army officer Clement Attlee did not seek to further the interests of the poor when he established the NHS and the welfare state as a consequence of the overwhelming mandate the Labour Party achieved in the 1945 general election?
LikeLike
The iron law is that the elected will be members of the elite. That is invariable. What is not invariable is the ideology that those elected hold. The tendency is for the elected to represent elite values and interests as opposed to those of the majority. But that is merely a tendency, to which there are exceptions. As can be expected based on principles of self-representation and as history shows, over the long term the exceptions get smoothed out and the tendency prevails.
LikeLike
Yoram,
Your point about the exceptions getting smoothed out over time is important. We say “absolute power corrupts absolutely,” but I think the more useful principle is that the corrupting effect of power varies with both the amount of power and the length of exposure to it.
Absolute power over a long period is extraordinarily likely to corrupt. A great deal of power over a long period is very likely to corrupt. A little power held briefly is much less likely to. There is tremendous individual variability, which is probably why only the extreme case became proverbial. You can always find people who resist those pressures unusually well. But over enough people and enough time, the underlying tendency should increasingly dominate the exceptions.
I think the same principle applies more broadly to institutions. Given enough time, I assume that essentially any institutional arrangement will degrade. Institutions are not fixed machines. They are more like living systems. The people within them adapt to their incentives and opportunities, while outside organizations and individuals actively adapt to the institution in order to influence, exploit, or capture it.
That is one of the basic assumptions behind my architecture. I don’t assume that power can or should always be eliminated. Sometimes substantial power is necessary to perform a function well. The design problem is to optimize the situation: remove power where it is unnecessary, constrain or compensate for its effects where necessary, and structure duration, incentives, selection, information, oversight and other environmental factors so that the desired behavior is favored for as long as possible.
But I still assume that adaptation and degradation will occur eventually. The objective is not to discover an institutional arrangement immune to them, but to design around that fact.
LikeLike
Yoram, I appreciate the critique.
Is there a good place for me to read a full explanation of your ideal sortition system?
As I think I indicated previously, my goal in using sortition is to most accurately approximately what [all the people who are expected to follow the rules] would decide if we could all fully participate, become informed, and deliberate towards consensus together. I’m going to summarize this as democratic accuracy. If you can show me a model that seems likely to be more democratically accurate, I will adopt it.
In the meantime, here is my defense for why I like Terry’s design, given that democratic accuracy is the most important criterion for me…
If you are willing, I would be very honored to have your critique (along with the critique of other theorists here) on the following three brief essays I wrote for Sortition USA. I am planning to refine them and turn each of them into long-form YouTube videos and I would like to improve them as much as I can before doing so.
Thanks!
Ian
Why Sortition?
https://open.substack.com/pub/sortitionusa/p/why-sortition?utm_source=share&utm_medium=android&r=6mdhb8
How To Use Sortition
https://open.substack.com/pub/sortitionusa/p/how-to-use-sortition?utm_source=share&utm_medium=android&r=6mdhb8
How We Win
https://open.substack.com/pub/sortitionusa/p/how-we-win?utm_source=share&utm_medium=android&r=6mdhb8
LikeLike
Yoram:> The iron law is that the elected will be members of the elite.
You have a very old-fashioned perspective on the elite (singular) vs the people. Manin only makes the (obvious) point that elections favour the distinctive. He doesn’t use the word distinguished and people are at liberty to choose the characteristics they prefer. And he doesn’t suggest that those chosen will privilege their own interests over their constituents.
Ian:> Creative, group-deliberative proposal crafting is a very complex process that depends heavily on what happens within the individual minds in the room and in what order those processes occur. It isn’t reasonable to expect to get the same output by repeating that kind of group creative process twice. But we can still perform validation by having another sortition sample compare the proposal against the status quo.
Many (most?) of us on this forum accept Terry’s argument for a distinction between proposers and validators, large short-term juries being the obvious solution for the latter role. But some of us disagree that sortition should be employed for both functions. Terry agrees that small group deliberation is vulnerable to problems of poor representativity and information cascades which his model addresses via a complex hierarchy of panels and groups. The legitimacy of such a solution would be hard to demonstrate, so some of us prefer the proposal role to be in the hands of political parties, but with a reduced threshold to ensure a better match with the discourses prevailing in society at any one time. This is partly for pragmatic reasons, but also to ensure that the demos has control of agenda-setting as well as the decision function.
LikeLike
Ian:> if we could all fully participate, become informed, and deliberate towards consensus together. I’m going to summarize this as democratic accuracy.
Although the original (Habermasian) theory of deliberative democracy valorised consensus, I believe this is no longer the goal in it’s current incarnation. The only examples I can think of consensus-based polities are totalitarian. If you want to introduce sortition into liberal democracies, you would be well advised to drop any reference to consensus. “Democratic accuracy” (which sounds like an empirical, rather than normative term) would need to acknowledge that agonism is an inevitable component and that there are no “right” answers to political decisions.
LikeLike
Hi Ian,
Thanks for responding and for engaging in an open-minded and reasoned discussion. I hope this will be fruitful for both of us.
You raise various points. Let me start with addressing a couple and I’ll address the others later.
> Is there a good place for me to read a full explanation of your ideal sortition system?
I don’t think that having detailed blueprints for an ideal sortition-based system is a good idea. For one thing it is an attempt at usurpation (see more on this here). The allotted system should be self-designing – a process that should be ongoing. Its structure should not be dictated by a self-appointed designer or by any outside force, which inevitably expresses an elite view. What we do need is a starting point for this process, i.e., an initial body which gets the self-design process going. Here is a post where I lay out parameters for the proper constitution of this initial body: part 1, part 2. To put things simply, the starting point for a sortition based system should be a body that is very much like a standard parliament/Congress, except that it is allotted instead of elected. That body should use its sovereign power to develop and modify the system as it sees fit, and the process should be ongoing so that structural modifications are introduced as necessary (including, perhaps, the creation of more allotted bodies and of division of labor between those bodies).
> my goal in using sortition is to most accurately approximately what [all the people who are expected to follow the rules] would decide if we could all fully participate, become informed, and deliberate towards consensus together.
Following Peter Stone, I don’t think this “mirroring” criterion makes sense – see here for my critique. The second part of the post presents what I think is a more fruitful way of thinking about sortition.
In general, democracy should be seen first and foremost as a government system that wins the approval of the members of the governed group. This definition, which is essentially what “consent of the governed” is about, is relatively straightforward to operationalize: people can be asked whether or not they approve of the government under which they live. (By the way, the fact that countries do not regularly do this indicates that their governments are not really aiming at being democratic.) The claim in the “extension of self-representation” theory of sortition is that the theory shows why sortition can be expected to generate a system which enjoys the consent of the governed.
LikeLike
Keith: I know “consensus” is problematic. My statement maybe needs to be reworked for clarity, but I am saying TOWARDS consensus to indicate that proposals should be tested and iterated to get as close to consensus as is reasonable without forcing the issue. I want to maximize the number of people who are satisfied by the outcome.
LikeLike
Yoram,
I appreciate you engaging with me. I think it will be a fruitful discussion.
I agree that it is not democratic to attempt to impose a preconceived sortition-based government system onto a population. But, I don’t think that the initial conditions for a sortition democracy will appear spontaneously, so reaching that future depends on getting a critical mass of supporters to coalesce around a reasonable first draft that can (and should) be iterated, improved, and expanded.
I strongly believe that the right approach is to come up with a robustly designed model for a repeating sortition-based constitutional convention. The most important goal is to get that sortition-based convention system in charge of the parts of constitutions that deal with amending and interpreting the constitution.
A group of us volunteers at Sortition USA has been working on trying to come up with a design for that first draft of a sortition-based iterative/repeating constitutional convention. I would love your thoughts on that!
Regarding “mirroring”
I agree with you that it is not actually possible within large societies to have all the governed “fully participate, become informed, and deliberate towards consensus.” The expansion of human cognitive and communication capacity necessary to meet this definition might result in an incredibly alien outcome. But that “if” statement could just as easily mean “as if” the society was small enough to make those conditions realistic. That’s what I mean. I want a comprehensible, relatable type of approximating the whole population (by scaling down the size of society), not an incomprehensible, alien version (by scaling up human cognition/communication).
The upshot of that last paragraph is that I agree with your “extension of self-representation” justification for the democratic accuracy of sortition when you say…
“A small group of people, under reasonably favorable conditions, is able to represent its own interests… Policy that promotes the interests of a small group of people which are selected as a sample of a larger group will tend to promote the interests of the larger group as well.”
If you have a suggestion for phrasing my “democratic accuracy” purpose in a more rigorous (but still rhetorically compelling way), I would be very interested.
By the way, I absolutely agree that a truly democratic society would include an institution to regularly ask the governed whether they consented (meaning deep, informed consent, not the ridiculous version of {assumed consent based on the lack of a successful revolution} that appears to be the current standard). In the constitutional convention design I linked above, we included a regular plebiscite question about whether to hold a constitutional convention early and open up the government to be replaced.
Help us make that convention design better!
LikeLike
Ian:> I want to maximize the number of people who are satisfied by the outcome.
In the parlance of liberal democracy this is generally operationalised as majority rule, rather than “moving towards consensus”. As I mentioned earlier, the latter trope (derived from deliberative democracy) is only in practice realised in totalitarian political systems. My preferred historical model is Athenian democracy, which was both majoritarian and agonistic. Given the uphill battle to implement sortition, it’s better to work within the existing terminology, rather than that derived from ideal (normative) theory, dismissed by Jeremy Waldron as “57 varieties of luck egalitarianism.”
I’m also surprised by appeals to public opinion surveys — Fishkin’s life work has been based on developing a way to measure informed public opinion, via an “attentive” minipopulus, so resorting to crude plebiscites would be a backward move. Of course you can’t have a minipublic deciding whether or not to have government by minipublics, hence the need for scholars to debate what sortition can and cannot do (both in theory and in practice) — Waldron’s “political political theory”. No doubt this will be dismissed as cow-towing to academic elites who are (according to Yoram) only in it for themselves.
LikeLike
Keith:
I’m not sure if majority rule fully captures what I mean. I want to be able to go beyond bare majorities when possible and try to find the outcome that will satisfy the most number of people without artificially requiring any specific threshold (I.e. supermajorities).
This can be accomplished by procedures such as…
I think that having a plebiscite on the question “do you think a constitutional convention is needed to reform the government?” is one of the few cases where a plebiscite makes good sense.
If many politically active people feel the government could be improved, it seems wise to attempt to improve it from a “consent of the governed” lens or even just a “prevent a violent revolution” lens.
The typical concern about plebiscites being a tool for those with the most capacity to manipulate public opinion doesn’t really apply here because this plebiscite would have no power to enact policy, only to convene a sortition-based constitutional convention to examine the reasons for dissatisfaction and attempt to mitigate them.
If the reason for dissatisfaction was elite manipulation of public opinion – rather than a problem with material conditions – the sortition-based convention could mitigate the public manipulation.
Similarly, the concern about plebiscites being subject to rational ignorance also doesn’t really apply to this use-case because we aren’t asking people to become informed on anything other than their own satisfaction with the existing government. No additional effort to become informed is expected or required. We are just asking people if they think a better governing system can be developed, based on their own experience.
Ian
LikeLike
Ian:> This can be accomplished by procedures such as…Choose the proposal that is acceptable to the greatest number using pairwise comparisons (I.e. condorcet voting methods)…Ask people to discuss proposals to surface the reasons people like them and don’t like them…Encourage people to develop variations of proposals in an attempt to improve the proportion of people that would find the specific proposal acceptable.
Sure, that sounds good, but “people” refers (presumably) to those participating in the deliberative exchange. My concern is always the overwhelming majority of citizens who don’t get to participate and would be (effectively) disenfranchised by the aleatory coup. Lottocrats claim that a stratified sample will ensure that they are adequately represented, but this is not the case if voluntarism is a statistically significant population parameter (my hunch is that it is at least as significant as age, gender, education, ethnicity etc). Add to this the tiny group size required for a meaningful deliberative exchange and the dangers of information cascades. If so then such a body, while fulfilling the normative requirements of equal speech etc can make no meaningful claim to represent those who don’t participate, hence the ongoing need for election for policy proposers (with a lowered threshold) and final decision making by large randomly-selected juries.
We should always remember that deliberative democracy originated as a normative procedural guide for intra-group exchange and was originally aimed at elite bodies (see Joseph Bessette’s 1994 book). That’s why I’m keen to distance the sortition movement from deliberative democracy (a view that I share with Yoram).
LikeLike
In his last book (2023), Habermas is surprisingly positive about agonism (he also acknowledges that consensus is only appropriate to a limited range of forums):
“The agonistic character of electoral campaigns, conflicts between opposing political parties . . . can be correctly assessed only if one recognises that the functional contribution of political mass communication . . . is that of generating competing public opinions on topics relevant for decision-making . . . Nothing more may be needed to spark an antagonistic dynamic within the public sphere than the conflict-generating orientation to truth that citizens associate with their expressions of political opinions. But that also tends to be functional in generating competing public opinions.” (2023, p. 71)
If that’s true for an ur-deliberationist like Habermas, I’m genuinely puzzled as to why lottocrats arrogate an agenda-setting role to citizens’ assemblies and other consensus-seeking sortitional projects.
LikeLike
Ian:> A group of us volunteers at Sortition USA has been working on trying to come up with a design for that first draft of a sortition-based iterative/repeating constitutional convention.
I have reviewed this rather intricate proposal. You are essentially proposing an amendment to the U.S. Constitution for establishing a rolling basis for further constitutional conventions based on sortition. I suggest you break this into discrete steps.
Step 1: Enabling legislation for calling a Constitutional convention. For example, sortition within each state to constitute a State convention with (say, 10 or 20 or X) members per electoral vote for that State. A time frame and subject area for the conventions to consider. For example, 3 months to consider possible amendments to Article V of the U.S. Constitution (procedures for amending the Constitution). Some voting procedure to aggregate the preferences of each State convention into a single set of possible amendment(s) to put to popular referenda in each State.
Step 2. Assuming the enabling legislation is triggered, i.e., by 2/3 of Congress or 2/3 of States within a specified time (e.g., three years), each State selects (randomly among eligible voters) potential convention attendees. Some method of compensation for (voluntary) attendance would need to be part of the enabling legislation. These initial conventions are disbanded on or before the specified time.
Step 3. Assuming these initial conventions approved one of more potential amendment(s), these would be referred to the people in national referenda. One method is for each state to vote between two slates of candidates sworn to vote either yeah or neah on each amending proposal. The winning slate becomes the ratifying convention for the state, which meets for one day to vote yeah or neah for that state. I believe this method was used for Amendment 21 (repeal of Prohibition).
Sortition USA (and anybody else) would be free to propose any changes to the future amending process. This comes under the heading, “Be careful what you wish for.” But good luck, if you can get more exposure for the sortition idea.
LikeLike
Keith:
I agree that voluntarism introduces bias that weakens accuracy. But using elections introduces even more voluntary bias than stratification to correct for small samples and drop-out. The pool of election candidates is voluntary and has much higher barriers to participation (access to campaign funding, free time, and a willingness to engage in public popularity contests). Voting in mass elections is voluntary and adds more bias (i.e. voting participation is closely associated with household wealth, education level, and other salient sociopolitical characteristics).
The question for me is: What is the most accurate way to represent the attitudes and preferences of the public in a body that writes the proposals for laws, knowing that effective group policy crafting likely depends on all-to-all communication?
From a statistical perspective, stratified sampling would be more accurate in cases where we have significant constraints on sample size (forced by the need for all-to-all communication) and drop-out rate (forced by the social acceptability/advisability of mandating participation for the prolonged periods necessary for crafting legislation).
Those constraints introduce weaknesses, but the voluntary, stratified random sample we would use to mitigate those constraints would still be the most accurate way to reflect the attitudes of the public at large in the proposal writing process, at least in my understanding of study design/statistics.
I am sympathetic to your and Yoram ‘s skepticism of deliberation.
I think that good sortition/random selection gives us accuracy and good deliberation gives us precision.
Precision without accuracy ensures that we consistently hit the wrong target.
Also, I think fetishizing deliberation behind credentialism and allowing a priestly class to gatekeep “good deliberation” opens up the whole movement for capture.
LikeLike
That being said, the best case scenario is precision AND accuracy.
It seems like there are good people and orgs within the deliberative democracy movement that aim to make well-considered (IMHO) procedures accessible and replicable without them.
LikeLike
Ian,
Your discussion of voluntarism suggests to me that you are conflating two distinct variants of representation — descriptive and the active representation of interests. We both agree that in the former case voluntarism introduces inaccuracy, but the latter variant does not require the representative to share the descriptive characteristics of her constituents. With election under universal suffrage people can volunteer to stand for office and also whether or not to vote for the candidates making a representative claim.
>: The question for me is: What is the most accurate way to represent the attitudes and preferences of the public in a body that writes the proposals for laws . . .
The answer for me would be election with a reduced proposal threshold. 50% + 1 leads to a choice between two unrepresentative tyrannies; drop the threshold and there would be a choice between a greater variety of candidates who would better reflect the ideological diversity of the full citizen body (and then have a large sortitional jury select the proposal that best matches their beliefs and preferences.)
>. . . knowing that effective group policy crafting likely depends on all-to-all communication?
Actually I agree with you, but the deliberation should be to seek an intra-party consensus. It’s up to the jury to decide which of the competing proposals best matches their own beliefs and preferences.
I don’t understand what you mean by “precision”.
LikeLike
Ian: Reading through this discussion, I feel as though the principles you are outlining for your ideal system mirror very closely the values I’m using when implementing my own system. A basic overview of the system can be read here: https://www.ingentaconnect.com/content/imp/jos/pre-prints/content-impjos9
Though why I believe this achieves the goals you outline will be presented in a future work (two papers away, I think) where I will discuss democratic legitimacy in the overall population. But, like you, I want the system to work for the greatest number of people, be constantly getting informed consent from the population in a deep way, etc. I believe (and will argue) that this is possible from the outlined proposal in the paper. If it grabs your attention, I’d be happy to discuss it in more detail in another venue.
LikeLiked by 1 person
Keith:
Your “active representation of interests” is a principal-agent relationship.
I just don’t think that principal-agent relationships are a more accurate way to get the well-informed policy preferences of the population than to get the well-informed policy preferences of a representative random sample (even one that is stratified to correct for relatively small size and voluntary drop-out).
Effective principal-agent relationships depend on the agent knowing the values of the principal(s) and the principal(s) having reasonable hope of holding their agent to account.
I’m very skeptical that politicians are prepared to accurately reflect the values of all of their constituents in a rigorous, balanced way on every issue they must act on.
I’m very skeptical that voters can effectively hold politicians accountable due to rational ignorance and the blunt tool of voting for an alternative, especially given that the pool of alternatives is limited to the voluntary pool of privileged, viable politicians.
Precision is the ability to hit the same spot consistently. Accuracy is the ability to hit the target.
Having a tight cluster of hits in one quadrant of a target would be precise, but not accurate.
Having a loose distribution of hits that centers over the bullseye would be accurate, but not precise.
LikeLike
Ian,
I share your view on the failure of principal-agent representation in the current incarnation of democratic politics. Our alternative proposal seeks to de-aggregate bundled policy preferences and reduce the proposal threshold sufficiently to enable a better reflection of the ideological diversity of the demos. Accountability is (over time) ensured by the need to secure favourable jury verdicts. We also propose a strict separation between legislative and executive functions.
>the well-informed policy preferences of a representative random sample (even one that is stratified to correct for relatively small size and voluntary drop-out).
You acknowledge the problems of small sample sizes and information cascades, and the impossibility of stratification correcting voluntary drop-off (currently around 95%).
From a pragmatic (implementation) perspective, turkeys (both politicians and voters) are highly unlikely to vote for Christmas. Much better to reform existing electoral practices and build on the public confidence in juries. This also happens to be a modern analogue of 4th century Athenian lawmaking, whereas lottocracy can only appeal to ideal political philosophy.
For a highly abridged summary of our position see https://www.academia.edu/44790587/Some_Problems_of_Citizens_Assemblies
LikeLiked by 1 person
Ian:> I’m very skeptical that politicians are prepared to accurately reflect the values of all of their constituents in a rigorous, balanced way on every issue they must act on.
Indeed, that would be an impossibility. Our proposal is a variant of proportional representation and a debundling of issues. This will result in smaller parties forming ad hoc coalitions over specific policy proposals.
>I’m very skeptical that voters can effectively hold politicians accountable due to rational ignorance and the blunt tool of voting for an alternative
Accountability is to the informed preferences of the jury. The number of policy alternatives (determined by the proposal threshold) would be a compromise between the need to represent heterogeneous discourses and the cognitive constraints of the jury (who would have to choose between them).
LikeLiked by 1 person
Ian,
> I want sortition bodies to only serve long enough to make a single well-considered proposal/decision. […] Having a single sortition body make many different decisions seems to unnecessarily risk amplifying any errors from the single sampling process.
First, as I indicated, the question is not whether you or I think that having single-issue bodies is a good idea. The question is whether the idea of sortition should be promoted as being tied to specific institutional designs. Determining a democratic institutional design should clearly be the ongoing job of an allotted body. If an allotted body decides to set up single-issue bodies in some contexts, that’s fine. However, promoting this idea apriori as part of the setup is an anti-democratic attempt to constrain the design by the allotted body.
Substantively, I find your argument far from convincing. Modelling politics as a sequence of independent one-off decisions is simplistic and counter-productive. Decisions need to be tuned over time and their interactions need to be taken into account. Having to make an up-or-down decision, and having to design a proposal with a view toward a one-time up-or-down decision is putting the system into a straightjacket. Again, it may be that this would work in certain cases, but seeing this as a general way to handle politics is in my opinion “all around a truly bad idea”.
Also, being concerned about sampling error is focusing on a negligible source of risk. For any sampled body that is larger than a handful of people, the risks lie very much elsewhere.
> Volunteers. I’m using this word to mean “voluntary.” I don’t mean that people could volunteer to serve.
If memory serves Terry is thinking of having bills drafted by people who actively expressed interest in a certain issue (but I may be wrong about this).
> Service would be by random invitation only, but we should attempt to minimize the proportion that decline by making service on a sortition body very attractive
Ok – with this I very much agree. Again, this is not current practice and I don’t think this is what Terry is proposing either.
> it is not anti-democratic to take another sample to perform data validation on the first sample’s work
There is nothing inherently wrong with different people looking at a proposal. There are advantages and disadvantages in the division of intellectual labor and in the division of authority. What is anti-democratic is the claim – that Terry makes and that you repeat (at 28:15 to 28:30, to be exact) – that people are poor judges of their own proposals. That is an argument you should avoid.
> I would be very honored to have your critique (along with the critique of other theorists here) on the following three brief essays I wrote for Sortition USA.
Sure, thanks, I’ll be happy to go over them. I’d also be happy to participate in a discussion where we discuss our differences as well as common ground. Personally I find discussions where people challenge each other to be more stimulating than sermons. I think I am not alone in this.
LikeLiked by 1 person
Yoram:> Determining a democratic institutional design should clearly be the ongoing job of an allotted body.
According to Gatocracy this is tautologically true, but all democrats (and most lottocrats) would dismiss such an all-powerful body as an aleatory oligarchy. I don’t recall anyone on EbL endorsing your claim.
LikeLike
Ian:
I’m not sure I understand your claim here. To be clear, the stance is that by investing time and effort into the creation a proposal, the person who writes it is too close to judge whether or not it truly has merit, or if they’ve simply talked themselves into thinking so. Therefore, someone other than the writer should be needed to evaluate it.
If the people who are deciding are picked in a democratic way, what would make this anti-democratic?
LikeLike
Jordan, I think this observation was from Yoram, not Ian. To most of us it’s self-evidently the case that nobody should mark their own homework.
LikeLiked by 1 person
Apologies, I was very aware it was Yoram. I’m unsure why I wrote Ian. I blame this on insufficient coffee consumption.
LikeLike
Jordan:> the person who writes it is too close to judge whether or not it truly has merit, or if they’ve simply talked themselves into thinking so.
In addition to James Madison’s claim in Federalist 10 that “a [single body of men are unfit to be both judges and parties at the same time”, this could be taken as a broader critique of deliberative democracy as currently practiced. According to Mercier and Sperber’s “argumentative theory of reasoning” (ATR), confirmation bias is a valuable cognitive attribute for policy proposers — to paraphrase E.M. Forster “how do I know what I think until I [hear] what I say.” The ATR would suggest that the converse ability (to judge between proposals) is ideally suited to a large silent jury. Deliberative democrats (and lottocrats) could learn a lot from Andrew Dobson’s book Listening for Democracy.
LikeLike
Jordan,
> To be clear, the stance is that by investing time and effort into the creation a proposal, the person who writes it is too close to judge whether or not it truly has merit, or if they’ve simply talked themselves into thinking so. Therefore, someone other than the writer should be needed to evaluate it.
So when you decide you want to do something, say, go into a certain profession, marry a certain person, or buy a house, could it be that this decision does not truly have merit but rather you simply talked yourself into thinking so? Could it be that you therefore need someone else to evaluate it for you, and have veto power over whether this decision is actually carried out?
LikeLike
Ian,
I had a look at “Why Sortition”. I think that the arguments need to be tightened considerably. For example, at the outset you discuss why normal people do not get elected. You offer two reasons. The first is because they have responsibilities they cannot abandon and the second is that they don’t have money to advertise themselves.
Which of those problems do you believe is the more important one? To me it obviously seems like the second. Although even the matter of money is not really the heart of the problem.
It does not take extreme privilege to be able to become a candidate. The real problem is becoming a credible candidate, which requires becoming known to a large portion of the voters. The voters cannot select “the best” of the entire population but only “the best” of the ones they know. This is inherently competitive since due to congnitive constraints (limiting total number of people known to any single person) only very few people can become known to a large portion of the voters. Thus some mechanism has to determine who becomes a credible candidate and this mechanism cannot depend on the choices of the population. The question of money is thus a secondary one, not a primary one. It is merely one possible mechanism for determining viable candidates. If it is somehow eliminated, a different mechanism would have to replace it (e.g., being famous to begin with, or a co-optation mechanism within political parties).
If you find this useful, I’d be happy to collaborate with you more closely on developing an effective presentation.
LikeLiked by 1 person
Yoram:>So when you decide you want to do something, say, go into a certain profession, marry a certain person, or buy a house, could it be that this decision does not truly have merit but rather you simply talked yourself into thinking so?
There is something to be said for arranged marriages (the default until comparatively recently)
>Could it be that you therefore need someone else to evaluate it for you, and have veto power over whether this decision is actually carried out?
The problem is when a random sample takes a decision on behalf of the whole population. As the proposing sample may well be prone to confirmation bias, it is better for its decision to be reviewed by a different sample (or for an alternative method to be used for selecting proposers). This problem is acknowledged by Madison (Federalist 10) and most lottocrats (including Bouricius), although it is anathematised by Yoramocracy. As always we need to shift our focus to the vast majority of citizens who will be disenfranchised by the aleatory coup, not just the tiny number who win the golden ticket.
LikeLike
Yorum:
These are personal decisions made for you, by you, that don’t affect anyone else. You know yourself better than anyone, you can be made to understand the risks, the commitments, and in the end, any bad decisions, the consequences rest with you.
With policy, it is not always going to be the case that the author knows the situation better than anyone else, as policy is something that touches many people in many different ways. They may only be viewing it from one particular perspective, their own. We may think we are creating a policy that works for everyone, but only because we can only see it from a particular set of perspectives. The very act of building it up, as well, creating reasons why this plan is a good one, makes us more and more convinced we are right. Confirmation bias helps us find those reason easily, while we fail to find reasons we might be wrong.
Look at all of us here, for example. All of us on this site, with a few exceptions, all have our own ideas on how sortition should be implemented. We are all here arguing in favour of our own designs and plans, and we are all emotionally attached to them, finding many reasons why they could work. We might be able to find some reasons why it might fail, but the reasons we come up to counter them may feel strong to us, until we have to provide those reasons to someone else.
Having others evaluate it, especially considering this policy is going to have an effect on others, ensures that someone who isn’t emotionally attached to it, didn’t come up with reasons to defend it through confirmation bias (something that affects us all), provides powerful oversight. Having a diverse set of people in that final group ensures the reasons provided can convince a diverse set of perspectives.
LikeLike
Jordan:> Look at all of us here, for example. All of us on this site, with a few exceptions, all have our own ideas on how sortition should be implemented. We are all here arguing in favour of our own designs and plans, and we are all emotionally attached to them,
Actually it’s only a tiny number of people who post their own ideas (and are subject to confirmation bias as a result). The majority of readers remain silent — they should be the ones to judge which of the ideas being floated are most plausible. How to obtain that aggregate judgment is another matter. And, of course, readers of this forum in no way approximate the broader demos — they are no more representative than the 5% of those selected by sortition who accept the invitation.
LikeLike
Keith:
I have expressed that this is where we disagree, and my next paper will take this on directly. It’s unfortunatley slow going, personal life has been busy, but look forward to a thorough discussion of this point in the future.
LikeLike
Jordan,
> These are personal decisions made for you, by you, that don’t affect anyone else. You know yourself better than anyone, you can be made to understand the risks, the commitments, and in the end, any bad decisions, the consequences rest with you.
When it comes to the argument that you made above, the question of whether a decision affects only its maker or others as well is neither here nor there. Your argument was:
> by investing time and effort into the creation a proposal, the person who writes it is too close to judge whether or not it truly has merit, or if they’ve simply talked themselves into thinking so.
If this argument is accepted, then people are unable to properly judge their own ideas. Therefore it would be better for people – for their own good – to have their own intended decisions judged by others, with the power to impose the judgment on the person intending to make the decision. This should hold no matter whether the decision is private and affects only that person or affects other people as well.
LikeLike
Yoram
This would assume that there is only one value we should be holding onto, regardless of scenario. Good decisions often need to be balanced against agency. As an individual, I want the capacity to make bad decisions, and I am willing to take ownership of those decisions if they don’t turn out right.
For governance, we must respect the fact that decisions we make are not just an act of our own agency, but impacts the agency of others.
But yes, if you are asking whether confirmation bias affects us in our own lives, the answer is “of course it does”. We absolutely make decisions in our own life with confirmation bias. We will have an instictual gut reaction to what we should do, and then look for reasons to back up that decision, often generating poor counter-arguments. Talking to someone else about the decisions we make and geting feedback often helps us produce better decisions in our own life.
However, as individual agency is something we value, to an extent, we are willing to accept allowing bad decisions in people’s lives. However, government, we should wish to be a bit more cautious.
LikeLike
Jordan,
> Good decisions often need to be balanced against agency.
So, do I understand correctly that you are saying that the need for agency is generally in competition with good decisions?
Are you saying that people who have their life decisions judged by others, and vetoed when those others see fit, generally end up with better life outcomes then those who are able to carry out their own plans?
LikeLike
I don’t know if I would make a full generalization. However, the need for agency is definitely in competition with good decisions *frequently*.
If we were to wish to find a way to always make the best choice in our life, I would argue that we would definitely include the input from others frequently, yes. Even better, groups of people who could deliberate and come to some agreement amonst themselves.
Though, one *could* argue that this would result in a bad outcome for me purely on the basis that agency is integral to “good” outcomes, inherently. If I’m not making decisions for myself and am just carrying out the decisions made by others, then my life no longer becomes my own, and this is inherently bad.
Nonetheless, there are elements of this that don’t survive analogy. How I decide decisions in my life and how legislation is written, which makes decisions for *other* people, will clearly need to have different considerations.
LikeLike
Jordan,
> the need for agency is definitely in competition with good decisions *frequently*.
That’s an unexpected stance. It seems highly anti-democratic to me. In my mind the principle that people are the best judges of their own interests is *the* democratic princple.
> If we were to wish to find a way to always make the best choice in our life, I would argue that we would definitely include the input from others frequently, yes.
That is a truism. We are all in favor of including input from others when making our own judgment. The question is about those others imposing their judgment on us.
LikeLike
Yoram:> In my mind the principle that people are the best judges of their own interests is *the* democratic princple.
Is that a normative, empirical or axiomatic claim? My understanding of democracy is that the people have power — how they choose to exercise it is a different matter.
LikeLike
> Is that a normative, empirical or axiomatic claim?
Sutherland! I believe this is the best question I have ever seen you ask!
I presume that you are referring the claim:
“People are the best judges of their own interests.”
I think that this is not a normative claim, since this is not a “should” or a “good/bad” statement. But the question of whether this is an axiom (or otherwise analytical) or an empirical (synthetic) claim is more difficult to answer. If it is not an axiom, how would you go about trying to show it is true or false (whether via pure logical derivation from other axioms or using a combination of derivation and observations)?
LikeLike
I think it’s axiomatic in both your example and mine. In my case it’s a matter of etymology (demo-kratia), in yours its the foundational principle of Yoramocracy. As to the evidence as to whether or not people actually are the best judges of their own interests, it’s a decidedly mixed bag.
LikeLike
> As to the evidence as to whether or not people actually are the best judges of their own interests, it’s a decidedly mixed bag.
So you believe there is empirical evidence for or against the “democratic claim”. What is the nature of this evidence?
LikeLike
People vary in their ability to judge what is in their own interest (but this has nothing to do with democratic claim making). I do agree that the aggregate judgment of large randomly selected juries is the democratic way to decide between competing representative claims. The demos is a non-reducible collective entity, so the methodological individualism involved in your axiomatic reasoning doesn’t apply. I’m currently reading the proofs of Daniela Cammack’s book on Athenian democracy and she makes this point strongly — the demos is constructed when the people assemble en masse.
LikeLike
> People vary in their ability to judge what is in their own interest
What is the type of evidence that can support this claim?
It seems that in order to make such a claim you would have to have an accepted “ground truth” criterion for what is in someone interests. What is that criterion?
LikeLike
It would be hard to argue that it was in anybody’s interest to consume excessive amounts of food, alcohol and addictive drugs.
LikeLike
> It would be hard to argue that it was in anybody’s interest to consume excessive amounts of food, alcohol and addictive drugs
That’s rather weak.
For starters, what does “excessive” mean? In reality, consumption of any amount of alcohol is bad for one’s health. Would that mean that any consumption of alcohol is against one’s interests? When do the costs of the long term health effects outweigh the benefits of the short term pleasure?
LikeLike
People tend to use common sense (rather than ground truth criteria) when judging these matters. I don’t have much interest in these matters, as they have nothing to do with democracy per se, the topic of this forum.
LikeLike
> common sense
To conclude: your scientific claim is that those who don’t conform to your common sense are unable to take care of their own interests.
LikeLike
commonsense (like democracy) is an aggregate term.
LikeLike
> commonsense (like democracy) is an aggregate term.
Who gets to be included in this aggregation? The same people who can’t take care of their own business?
LikeLike
Yoram, would you be able to elaborate for me your perspective of rationality and democracy ? Are all actions people choose rational? Is it a democratic principle that this necessarily be true?
LikeLike
Jordan,
As I wrote, it seems to me that the basic democratic assumption is that each person is the best representative of their own values and interests. That does not mean that every action of every person is rational but that most actions of most people are rational.
LikeLike
Yorum, let’s separate fundamental operational values. I want my house to remain at a comfortable temperature. If my air conditioner breaks, i am perfectly able to express my rational desire for a functional air conditioner.
This does not mean I know the best air conditioner, how to install it, what makes sense for my home, etc. The “how” is just as important as the “what”. Many disagreements in society are not fundamental values. We all agree that we want less crime and better healthcare . What we often disagree on is the instrumental values. People disagree on the best way to achieve these fundamental values. This isn’t all disagreements, but it’s many.
If people are arguing in favour of an instrumental value that all evidence points to being completely incapable of achieving the fundamental value, while a different option would achieve it better, I assume you would agree that this could be described as “irrational”.
LikeLike
how I wish we could edit. That first sentence should read “let us separate fundamental and instrumental value”.
LikeLike
Jordan,
I am not sure I understand what this distinction is about.
Sure. In many cases it is convenient to talk about means and ends. The democratic assumption is that people are generally able to judge both. Of course, often the means to an end is getting the help of others. But it is still people who decide to ask others for help rather than the others who impose their “help” on those whom they judge needs it.
LikeLike
Yoram:> Who gets to be included in [aggregate common sense]? The same people who can’t take care of their own business?
Sure, that’s why it’s called the wisdom of crowds (or, as Thomas Paine entitled it Common Sense). The fact that people can take care of their own business does not (necessarily) apply to individual persons — if it did then there would be no need to invoke the law of large numbers. Democracy and commonsense are aggregate terms — to invoke them using the language of persons is a reductionist category error.
LikeLike
Yoram, certainly you don’t believe that all people can rationally decide the means needed to achieve their ends? That would require quite a bit of niche and nuanced knowledge from everyone. I can certainly attest that I lack such knowledge.
LikeLike
> wisdom fo the crowds […] the law of large numbers
Aha. So can “common sense” – this unimpeachable standard of rationality – be determined using opinion surveys?
LikeLike
Jordan,
> certainly you don’t believe that all people can rationally decide the means needed to achieve their ends?
I certainly do. Again, to clarify, this does not mean that they make this decision in isolation without the help of others. People often (solicit and) willingly accept (even pay for) help from others. If they don’t, “help” imposed by others is generally going to be detrimental.
> I can certainly attest that I lack such knowledge.
What kind of knowledge are you referring to? Is this referring to specialized knowledge like that of a doctor?
Obviously people do not possess by themselves every piece of specialized knowledge. When I want a medical check-up I don’t check myself up. I find and avail myself of the services of a good doctor. I then listen to the doctor’s advice and most likely follow it (or get a second opinion). But I certainly do not allow the doctor to impose treatments upon me. Ultimately, it is my decision what I do. I determine both the ends and the means. Would it be better for me if this were not the case?
LikeLike
>Aha. So can “common sense” – this unimpeachable standard of rationality – be determined using opinion surveys?
That would depend on the topic (and how the question is phrased).
LikeLike
Unsurprisingly, the plot thickens. Who gets to determine how “common sense” is to be ascertained on each topic?
LikeLike
But Yoram, in a democracy, somebody else *is* deciding for you. Perhaps you may choose to reject a medical procedure based on a gut feeling, and perhaps you can argue that guy feeling is right for you. Perhaps you have a fear of needles, and working through that fear is far more challenging that dealing with whatever condition you currently have.
But in democracy, decisions are not being made just for you, but for all citizens. Perhaps you can take the perspective that any choice a person makes is the right one for them, but how can you extend this to making the right choice for others?
LikeLike
Jordan,
> Perhaps you may choose to reject a medical procedure based on a gut feeling, and perhaps you can argue that guy feeling is right for you. Perhaps you have a fear of needles, and working through that fear is far more challenging that dealing with whatever condition you currently have.
Attributing to others hypothetical problematic psychological constructs (“fear of needles”) in order to explain differing opinions or actions in a given situation continues the classical elitist, anti-democratic line of thought you have been arguing for. The self-assurance that you, or anyone else, can determine with high confidence that other people are behaving against their own best interests follows naturally from that tradition.
In fact, people may have different assessments of the physical or social facts. You, for example, may believe that following the doctor’s advice would clearly promote your health and would follow that advice without a second thought. Other people may believe the medical system is distorted by the profit motive or by other social forces and thus see it with more suspicion.
What is the objective criterion by which you can decide that one view is rational while the other is not? Without such an agreed criterion, you are simply asserting the superiority of one opinion over others.
LikeLike
>What is the [objective] criterion by which you can decide that one view is rational while the other is not?
The clue is the adjective in common sense.
LikeLike
Yoram, I’m agreeing with you. Different people may have different assessments about what is in their own best interests. The person who has a fear of needles has to legitimately take that into account when they decide upon a medical procedure. There is an increased trauma related to needle-related procedures that wouldn’t exist for other people.
Meanwhile, another person who sees the medical industry as being profit motive may, in fact, make different decisions than someone who doesn’t. I’m not denying any of this, or pointing to a “common sense”, something which I have great skepticism around as well.
What I am saying is two things. The first, is that it is possible to convince something is true when it isn’t. For example, let’s say that someone believes the medical industry has a significant profit-motive, and thus, many medical reccomendations are, in-fact, unnecessary but are pushed for profit reasons. Heck, let’s assume they are correct and take this as fact.
This will mean that there will exist some medical procedures that are not necessary but are reccomended anyway, while there will be other cases in which they are entirely necessary and not being pushed for profit motivated reasons.
Let’s pretend we are an outside observer with omnicient powers to know which is which. We would observe examples of both people accepting medical advice they do not need, and people rejecting emdical advice they do, based upon their personal beliefs and biases. Both of these people will be taken actions that go against their best interests based upon being misinformed. They will be *attempting* to achieve the correct ends, but will be mistaken about the means.
The second point I’m making is that different people will disagree with one another. However, in the case of policy, one decision ends up being applied to all. When creaating policy to acheive some ends, some specific means has to be selected, and if people disagree on which means are the best one, some portion of the population may be unhappy with that outcome.
Is there anything here you disagree with?
LikeLike
Jordan,
> Both of these people will be taken actions that go against their best interests based upon being misinformed. They will be *attempting* to achieve the correct ends, but will be mistaken about the means.
Sure, people can and do make mistakes and regret them later. But in fact, it seems to me that there is no useful way to define mistakes other than as something that a person regrets. The appeal to the hypothetical all-knowing observant is an erroneous and misleading way of analyzing the situation. (By the way, even the regret criterion is somewhat problematic because people’s values and interests may change over time.)
> if people disagree on which means are the best one, some portion of the population may be unhappy with that outcome.
Sure. But I don’t see the connection to the first point.
I am trying to keep the discussion focused on the question that started it – Terry’s argument (which you endorsed) about people being unqualified to evaluate their own proposals due to some sort of bias (i.e., a cognitive defect). As far as I can see, the disagreement between different people is irrelevant to that. Terry’s argument is not based on disagreements between people but about some supposed process of distortion that people undergo as they come up with proposals. My argument was aimed to show that Terry’s argument rests on anti-democratic assumptions about the incompetence of people.
LikeLike
Yoram, when it comes to knowledge of the means, do you believe that there exist things that are true or false, even when people believe those things to be otherwise? If someone thinks that arsenic improves fine motor skills, and then dies without regret, did they take an action that goes against their best interest due to misinformation, or would you suggest they were entirely correct as they had no regret (mostly because they were too dead to have regret)?
I don’t think it’s elitist to suggest there is true and false information out there, and that people can become confused on which is which. I have been misinformed on a great number of things in my life, and I’m certain I continue to be misinformed on a great of other things. Would you not say the same for yourself?
LikeLike
>Terry’s argument is not based on disagreements between people but about some supposed process of distortion that people undergo as they come up with proposals.
Terry (like Jordan and myself) relies on the argumentative theory of reasoning (ATR). This has nothing to do with incompetence, just the existence of different cognitive modules to persuade (both oneself and others) and to evaluate. The ATR is derived from evolutionary psychology and has no normative entailments — “corruption” being used in a purely descriptive sense.
LikeLike
Keith: The argumenative theory of reasoning is not how I justify that separation, and I actually disagree with the interpretation of it in that way. In the writing of legislation, it is entirely possible that both the proposing of arguments and the evaluation of them are used, provided the group who creates them contain a variety of different perspectives. Extending the “proposing” module and “evaluating” module to groups of people extends the theory in a way that I would argue isn’t supported.
However, the way your model and my own are constructed does not guarantee that the creating of legislation is done through a diversity of perspectives, and in the case of having several competiing proposals, there exist external motivations that can interfere with the deliberative process.
LikeLike
Jordan,
I wasn’t actually referring to democratic policymaking, only the ability for individual persons to decide what is in their best interests (the topic of Yoram’s question). We are all subject to confirmation bias and other cognitive pathologies in our everyday life and the best way of addressing this is generally COMMON sense (and accumulated wisdom).
LikeLike
Right, but the overall discussion is the separation of proposing and deciding of policymaking. Yoram thinks this is inherently antidemocratic, which is definitely an interesting position to take, and you and I agree that there exist reasons that those who write a policy may not be the best people to decide if it’s the best one. However, you included me amongst the people who use the AToR for that basis, so I wanted to clarify my own position.
LikeLike
Sorry Jordan, I thought you referenced the ATR in your paper, but I must have misremembered.
LikeLike
No worries, it’s a minor point at the end of the day. I do reference it, but I use it for other purposes, not the separation of bodies.
LikeLike
Jordan:> provided the group who creates them contain a variety of different perspectives,
I don’t see why variety per se has any direct connection with democracy. What matters is that the proposers match the ideological diversity of the citizen body and that the threshold should be lowered sufficiently to ensure that the proposals considered by the jury match this diversity (given the limitations imposed by the decision process). Proposers will combine with others sharing similar perspectives to their own. The fetishisation of diversity is a product of the epistemic corruption of diversity, and the sanctification of the Scott Page theorem beyond a simple mathematical truism. If we bang on about diversity too much the ongoing backlash against DEI might well kybosh sortitional democracy.
LikeLike
The argumentative theory of reasonsing depends upon a diversity of perspectives. The whole underlying concept of the theory is that the strength of reasons provided are proportional the the need to actually convince people. A group of people who all have the same views will produc weak arguments, as no effort is needed to convince anyone of the correctness of the views they already agree upon. A diverse set of perspectives will produce strong arguments, as those that produce arguments will require a higher threshold needed to convince people.
The best way to avoid group think, therefore, is to ensure diversity of thought. This isn’t my argument, but the argument of Mercier. This is made explicit in Reasoning Is for Arguing: Understanding the Successes and Failures of Deliberation from Mercier/Landemore.
If a group has diverse perspectives and the group is tasked with producing legislation, that task will require various members of that group to be, at various times, both proposers and evaluators of argument. The creation involved the production and evaluation of many arguments, the quality of which are predicted by the ATR to depend on the diversity of thought during creation.
The ATR does not suggest, as far as I can tell, that groups of people, sufficiently diverse, would be unable to evaluate the quality of the proposal they produce compared to a different diverse group, purely on the basis of being “proposers”. We might predict they would produce higher quality proposals based upon the knowledge they will need to *convince* another group, rather than only needing to be convince themselves. This is, for example, how Mercier and Sperber propose “solitary geniuses” operate.
LikeLike
Jordan,
> when it comes to knowledge of the means, do you believe that there exist things that are true or false, even when people believe those things to be otherwise?
Of course I believe there are things that are true and others that false. I just don’t think my beliefs should be privileged over other people’s beliefs. Again, it is a classic elitist notion that some people should be able to imposes their “true” beliefs over other peoples “false” beliefs.
> I have been misinformed on a great number of things in my life, and I’m certain I continue to be misinformed on a great of other things. Would you not say the same for yourself?
Indeed I have, which is exactly why I have become much less inclined toward imposing my beliefs on others. However, it is clear to me that, however often people are mistaken, most people are the best representatives of their own interests most of the time.
Somewhat of an aside, but relevent: Since you say that you changed your mind on a great number of things, what are the most significant things about which you have come to change your mind?
LikeLike
Jordan,
> the overall discussion is the separation of proposing and deciding of policymaking. Yoram thinks this is inherently antidemocratic
That is not my position. What is inherently anti-democratic is Terry’s argument for the separation, not the very idea of separation. Separation of proposing and deciding, like any division of labor, could have both advantages and disadvantages. I suspect that it is overall not a good idea (although that may also depend on the details of the institutional arrangements) but I don’t see it as inherently anti-democratic.
Also, promoting sortition on the basis of this specific arrangement is anti-democratic in the sense that it is an attempt to impose constraints on the institutional design. The institutional design should itself be an ongoing task for an allotted body without constraints that are imposed on it from the outside.
LikeLike
Yoram:
Isn’t this what the outcome of any “policy” is, the decision that *this* is the correct policy, and everyone has to live under it? What form of government would not result in the imposition on the citizens?
LikeLike
Jordan,
Again, the point in question is Terry’s argument. This argument makes a claim about people’s cognitive failures in general. It is not about government policy.
LikeLike
I mean, I’m confused here. He’s making the argument that confirmation bias might be present in the creation of legislation amongst the people who write policy. Confirmation bias is a well-known and studied element of human psychology present in all people, including the “elite”. Terry isn’t suggesting that there exists an “elite” who lack confirmation bias. The people who decide in the end are randomly selected citizens who vote, not an elite group.
There’s no group of “elite” who know the means better than anyone else, so I’m not sure what you are getting at.
We could, all of us, agree that the form a government takes should be created through an allotted body without consraints imposed by the outside, and still advocate for specific designs. The allotted body would likely look to the literature and review what options exist, and they pick and choose various elements from different systems and combine them, use a known system outright, or create their own wholesale. No matter how they end up doing it, having more arguments in favour of various systems available can only be beneficial to their process. Building up the arguments for a democratic body to make use of isn’t anti-democratic.
LikeLike
Jordan:> A diverse set of perspectives will produce strong arguments, as those that produce arguments will require a higher threshold needed to convince people.
I’ve just re-read the ATR chapter in Helene’s 2013 book. Although the topic of the book is the (epistemic) value of diversity, in this chapter she only recruits the case for diversity to refute Sunstein’s law of group polarization (a specific pathology of deliberative democracy).
The ATR was developed by the evolutionary psychologist Dan Sperber(2000, 2001) partially as an attempt to understand why people find it so hard to change their own mind — his explanation being that confirmation bias is deeply wired into the homo sapiens brain as a survival mechanism.
We are all prone to confirmation bias, especially those with an axe to grind. This being the case, deliberative democrats (Landemore, Mercier, Rose etc) focus on the need to structure deliberative exchange in such a way as to encourage the changing of minds of the participants (even though the ATR would suggest this is a doomed endeavour), whereas those of us (Kovner, Sutherland, O’Reilly) who focus on the distinctiveness of the two cognitive mechanisms (persuasion and evaluation) claim the ATR would suggest two different bodies (persuaders and evaluators). Terry Bouricius is in the latter camp (we were co-authoring a paper on the ATR), but his sortitionist monomania leads him to reject the obvious solution for democratic proposal making.
If you want a really strong example of confirmation bias, then read anything by Yoram Gat. My deliberations with Yoram span some two decades but he hasn’t shifted his views one iota. This confirms my view that popular sovereignty requires that decision makers should not participate in speech acts as (paraphrasing Madison), nobody should get to mark their own homework — especially those who get to set the questions. What matters is the (informed) view of the silent majority — note that Helene acknowledges that most jurors in the Bloomfield Track DP claimed that it was the information (not deliberation) stage that determined the outcome:
“The simple process of jurors seeing the site for themselves, focussing their minds on the issues, and listening to what [competing] experts had to say did all the work in changing jurors’ attitudes. Talking among themselves, as a jury, did virtually none of it.” (Goodin 2008: 58-59)
LikeLike
The ATR does not suggest that changing of minds is a doomed endeavour. It suggests that it is *challenging*. People will not change their mind without very strong reasons to do so, and will demand very strong reasons. Argument production is taylored to chanage minds, and becomes stronger the more effort is needed to do so.
There are, however, some catches. People who are externally motivated are unlikely to change their minds (as they aren’t actually looking to make the best choice, but the choice that is being externally motivated). Also, people need good reason *to* change their mind. If they don’t actually need to make a decision, they generally don’t need to actually change their mind.
LikeLike
My experience on this forum is that arguments have little effect on people who have already taken a strong position. Or are you suggesting that (e.g.) Yoram, Terry and myself are “externally motivated”? I take that to be a euphemism for having skin in the game. If so, what are the sinister influences pulling our strings?
>: Also, people need good reason *to* change their mind. If they don’t actually need to make a decision, they generally don’t need to actually change their mind.
I take a more charitable view regarding the pursuit of truth as an end in itself.
LikeLike
I don’t think you are externally motivated, but intrinscally motivated. Having skin in the game means you are motivated by the outcome itself.
By externally motivated, I mean that your job would depend on what final answer you produce. If changing your mind may cause you to lose your job, then your decision will be less likely to be based on the actual outcomes of the decision, and more based upon those external motivations. For example, if someone is being bribed, that’s an external motivation. Any benefit they get isn’t because of the consequences of the choice itself.
Politicians are generally highly externally motivated. The future of their electoral campaigns, funding, and party status will frequently depend how how they make choices.
However, the group of us here on these forums are a classic example of confirmation bias without a final choice to be made. The ATR offers a pretty good picture of why this might be. There is a tough problem to solve (how to make good government). All of us could consider all possible options, and look for the best arguments in favour and against each and every one of those options, but that ends up eing highly inefficient.
Instead, each of us forms an intuition, looks for evidence to back up that intuition, and create the best possible arguments for their own version of a soution. If ever there were a final decision to be made, say, everyone on the forum was tasked of coming up with a proposal that a government will actually use, I would imagine we would all be able to come together as a group, argue for a few days as to what this should look like, and come to some degree of consensus that most of us could live with. All of the arguments we’ve devloped separately would come into play, but the timeline of needing to produce an actual decision would change the dynamics and make us settle and compromise. It’s possible that we might even leave, all of us still convinced that our own model would be better than the group model we produced, but that the group model was “second best”, and better than all of our competitors. We may find ourselves satisfied with that outcome.
We may not have convinced one another that we were “wrong”, but the output would require us becoming convinced what “good enough” looks like.
LikeLike
>: We may not have convinced one another that we were “wrong”, but the output would require us becoming convinced what “good enough” looks like.
Yes, I can understand the dynamics of such a process. The trouble is there’s a lot of “us” and “we” involved, and that begs the question of whether or not the compromise thrashed out behind the doors of the smoke-filled room still reflects the considered preferences of the target population. Rather than a diversity of individuals it would be better to have a diverse range of parties producing a variety of options for the jury to choose between. Whenever there is talk of consensus I start counting the spoons.
LikeLike
I agree with being concerned with the broader population, but that’s a separate note. We are discussing what the ATR does and does not imply. The ATR simply suggests that the strength of argument production is related to the need to convince others. That people, reasoning on their own, will tend to convince themselves that their intuition is correct (confirmation bias). However, groups, reasoning together, with a diverse set of opinions, tend to produce stronger arguments and more valid output.
My point is that taking this argument and applying it to groups of people doesn’t follow. It is not correct to say that a deliberative group who produces a proposal (the proposers) have, collectively, only used the “production” module, and in order to determine the strength of the proposal you need a separate group to engage purely in the “evaluation” module. The group, when creating the proposal, will be using both modules during the production.
LikeLike
Jordan
> I’m confused here.
It does indeed seem to me that at least one of us is confused. My impression is that through a fairly long exchange we have been pursuing a certain line of argument. I was arguing that Terry’s claim rests on anti-democratic assumptions (and you seemed to agree that those assumptions are indeed implied by Terry’s argument but maybe felt that those assumptions are not problematic).
Now, you change direction and completely ignoring that previous exchange you are offering a completely new argument (namely, that Terry’s claim is supposedly established empirically). Such an abrupt change in the line of argument seems a rather problematic tactic if we are aiming at a fruitful conversation.
(As for that new argument: Whatever is “well-known” about human psychology, I doubt that it would be a blanket statement that would justify Terry’s claim, that people are unable to competently evaluate their own ideas. [Wouldn’t that imply that the more we think about something, the less confident we should be about our understanding of that something?] But since you seem so convinced that Terry’s claim is based on established science, maybe you can provide citations of the literature establishing the relevant findings.)
LikeLike
Yoram, I’m happy to provide sources on confirmation bias.
https://pmc.ncbi.nlm.nih.gov/articles/PMC9038198/
https://aisel.aisnet.org/icis2015/proceedings/HumanBehaviorIS/5/
That’s a couple of them.
But I don’t even think that’s fully necessary, as we agree that people are decent judges of their own fundamental values (ends) but rely on others for the instrumental values (means). This means that we require someone who is *trustworthy* to provide us the means, and there exist various parties who may have reason to deceive on what the best means are to achieve the ends.
But I’m still trying to work out what is “anti-democratic” about this notion. If we were to say these biases and problems exist for all people, without exception, what makes this claim specifically anti-democratic? Must we accept the claim that every single person must know all the best possible means for their interests? Or can we not suppose there exists a way for people, through the sharing of knowledge and working together in groups of some form, with checks and balances, can produce decisions that can work for the benefit of society, without every individual person being capable of doing it independently?
What is the definition of democracy you are using?
LikeLike
Sorry for the extra bit here, but also, what exactly makes a claim “elitist” if the proposition is that it applies equally to all people?
LikeLike
Jordan,
> That’s a couple of them.
From the abstracts these appear like random links under the heading of “confirmation bias”. Neither asserts (at not least in the abstract) that people are poorer judges of their own ideas than others are. (As I pointed above, it is quite unclear how such a claim could be substantiated empirically even in principle.) Seeking information that matches with existing beliefs could be useful for various reasons.
> people […] rely on others for the instrumental values
But if people regularly do, then the allotted would also do that. And then, what is the problem your arrangement is trying to solve? Why do we need to force the allotted not to make the final decision themselves (after having consulted with other people, if they wish)? It seems that you suspect most people will not actually behave as you want them to and thus you wish to imposes your ideas on them. (This answers your question in the post-script.)
That’s the standard Socratic self-contradiction. If we all agree that we need to hire a professional flutist or ship’s pilot rather than play or navigate ourselves, then why do we need to force people to submit to those professionals? If we all agree something is the right thing to do, we do not need to mandate it, since everybody would do it anyway.
> Democracy
Demcoracy is a society arranged in a way so that its members believe it is being managed according to their values and interests.
LikeLike
But Yoram.. The people who are evaluating are also randomly selected citizens using their own values, not mine. Do you not trust them to look out for their own best interests?
LikeLike
Jordan:> The group, when creating the proposal, will be using both modules during the production.
The producer group will be evaluating in a strategic manner, asking itself what reasons will be most likely to sway the jury. The task of the latter group is to evaluate which of the competing claims are right (in the epistemic/normative sense). If the producer group is seeking consensus they are utilising Habermasian communicative reasoning (the forceless force of the BETTER argument), which is very different from the dualism of the ATR. That’s why I’m puzzled that a deliberative democrat like Helene Landemore should utilise the ATR (and it’s why Terry Bouricius withdrew).
LikeLike
Keith:
I think Landdmore, as well of the authors of the ATR, have a different interpretation of the ATR. The primary argument of the ATR is not that there exists a dualism. Production and evaluation are things that work together, dynamically.
The theory is based on the interactions between the producer and the evaluator, and the role of producer and evaluator shifts back and forth during the deliberative process.
From Mercier/Landemore 2012,
In fact, the whole point here is that people are not good evaluators of arguments, per se, but that people are good evaluators of arguments they disagree with. The basis for which people are good evaluators of arguments is that people do not wish to be convinced by bad arguments. However, this only applies to arguments they don’t already agree with, as there is no need for htem to be convinced. When people agree, Mercier/Landemore argues that deliberation doesn’t even occur at all:
In this context, the evaluators are not good judge of arguments, as they are not skeptical of the incoming argument.
The ATR doesn’t posit that there exist separate modules that should motivate separate actions. It posits that these modules work best when interacting directly with one another, in a deliberative setting.
LikeLike
Jordan,
> The people who are evaluating are also randomly selected citizens using their own values, not mine. Do you not trust them to look out for their own best interests?
I trust people to try to promote their own best interests, but how effectively a body can promote its own interests depends on the details of the institutional design. A short-term body is inherently less capable of promoting its own interests than a long term body. A body that is constrained to an up-or-down vote on a given proposal is inherently less capable of promoting its own interests than a body that can come up with its own policy ideas. For these reasons, I don’t see any advantage in the proposers-deciders division of labor.
That said, maybe such an advantage exists and is yet to be effectively argued for. Terry’s argument for this division of labor, which is based on the supposed incompetence of people to judge their own ideas, however, is a very poor argument which is based on anti-democratic premises.
LikeLike
Jordan,
As I mentioned above, confirmation bias would predict that Sperber’s EP-derived theory will be interpreted differently, depending on the use to which it is being put. Landemore acknowledges that confirmation bias is a valuable trait in producers, but a deficit in evaluators. This would accord with the regular EP perspective on how different cognitive modules evolved, but is anathema to Rawlsian and Habermasian normative theory. The trouble with deliberative democrats who want to put a naturalistic gloss on a normative theory is they want to have their cake and eat it. Convincing others and being convinced are different functions and adopting an argument in order to persuade others (rather than pursue the truth) contravenes discourse ethics.
>: Production and evaluation are things that work together, dynamically.
Sure, and the best approach is the feedback loop between the two bodies — producers will seek to anticipate the reaction of the evaluators. That, rather than the pursuit of unanimity amongst producers, is what the ATR would suggest. The quote below, from a (forthcoming) JoS interview (by Baogang He) with Habermas is the deliberative alternative:
“During our meeting, I asked Habermas what should happen when public discussion fails to produce agreement. He answered without hesitation: “Then continue the dialogue.” I pressed him: “But what if disagreement still persists?” He replied seriously: “Continue talking until a consensus is found.” He then offered a detailed explanation: Any consensus must be grounded in mutual understanding and recognition among subjects. The binding force of communicative action arises when one party offers reasons intended to persuade others. If we are convinced by the argument, that party’s validity claim is thereby vindicated. By contrast, if the party resorts to violence rather than rational argument, it cannot acquire legitimacy.”
Accepting the most powerful argument is entirely different from privileging the forceful force of the BEST argument. Ultimately it’s an appeal to rhetorical violence, so this is the domain of Clausewitz, not Habermas. My own view (echoing Daniela Cammack’s forthcoming book) is that democracy was forged out of violence, so we need a correspondingly realist theory to improve it, rather than relying on normative hand-waving.
LikeLike
A good way to illustrate the incompatibility of Habermasian deliberation with the ATR is to compare DD and court-room practice. Deliberative democrats exchange reasons within a small group in an attempt to uncover the truth, ideally via consensus, so the confirmation bias is a pathology that needs to be overcome. The job of court-room attorneys, by contrast, is to advocate for their client’s interests by the use of persuasive rhetoric. Whilst the lawyer is not required to believe anything the client tells her, it probably helps if she does, so confirmation bias is a valuable cognitive asset when it comes to persuading the jury. The task of the jury is to determine the truth by deciding which of the competing narratives is most persuasive.
Although they share a common terminology, the two forms of deliberation have very little in common. If, as Cammack claims, the origins of democracy should be seen in the context of civil war, then the ATR/courtroom analogy is the most apposite.
LikeLike
To this we all agree. People do try to promote their own best interests, and how effectively they do that depends on the design. This is because humans have various limitations, and whatever design choices we make will have pros and cons depending on how those limitations interact with the deisgn.
One of those limitations are biases that can interfere with our ability to come to the best conclusions, even for our own self interest. The illistration of how confirmation bias interferes with this is fairly straightforward. If your initial instinct is that A is preferable, and that results in you looking for evidence that A is preferable, and don’t look for evidence that B would actually be preferable, then this actually does very little to demonstrate that A is, indeed, the better choice. What we generally do in science is if we believe that A is correct, we look for evidence that A is actually wrong, as failing to achieve that is far better evidence that A is correct.
However, confirmation bias has us pursue the worse strategy.
Having government design take this into account helps people make better decisions with regard to their own best interest. Again, this is true for *all* people, not just the non-elite.
LikeLike
Keith: Let’s forget, for the moment, what deliberative democrats argue. Deliberation is a process that can exist outside of their own arguments, and benefits can be demonstrated without citing a single deliberative democrat argument.
These aren’t in contradiction. We provide reasons to convince others, yes. Others, meanwhile, produce reasons to convince us. They evaluate our arguments, we evaluate theirs. The argument that is more persuasive is more likely to actually cause the other person to change their mind. Arguments that are factually untrue tend to be unpersuasive. At least, once it’s pointed out and demosntrated that it is untrue.
In the ATR, confirmation bias is a shortcut that groups of people can use to find the truth more efficiently. Each person, independently, produces the best argument from their intuition, and the strongest arguments survive the deliberative process and weaker arguments are discarded. People *are* convinced by better arguments. This last statement is fundamental to the theory. From the Enigma of Reason:
The theory doesn’t work unless people change their minds and accept good arguments.
I have not gone far enough into my readings to verify this assertion. However, assuming this is what they claim, we can diregard that and look at deliberation in of itself. I argue that confirmation bias is not a pathoogy that needs to be overcome, but the very reason group deliberation works in the first place. Confirmation bias brings in the strongest possible argument of various perspectives, and pits them against one another. It isn’t a pathology to be overcome, but the fundamental tool that powers deliberation entirely. Each perspective brings in additional information not found in the others, allowing the final conclusion to be stronger than all of the individual conclusions are independently.
LikeLike
Jordan >: The argument that is more persuasive is more likely to actually cause the other person to change their mind.
That certainly doesn’t apply to opinion silos (such as this blog). Once someone has committed themselves to a position it is very hard for them to change.
>: The theory doesn’t work unless people change their minds and accept good arguments.
That’s right, and both the ATR and jury trials would suggest that it is easiest for people who are agnostic on any particular matter to change their minds (not much movement is required). This agnosticism is best preserved by exposing people to competing arguments and then allowing them to exercise their free choice. Given that (unlike lottocrats and hardcore deliberative democrats) we agree that electoral competition is the best vehicle for policy proposals in a democracy, why do you feel the need for face-to-face deliberation? The wisdom of crowds suggests that people can make up their own minds on an issue, without the need to rely on peer pressure.
Your quote from the Enigma of Reason (which distinguishes between “audiences” and “communicators”) would suggest that the author agrees. Face-to-face deliberation adds no value and is antidemocratic (as the agents are persons rather than the demos).
LikeLike
Keith:
Well, no, the book says the opposite. In this case, audience simply means “someone who is listening to an argument”. This person need not be silent, though. If they speak, and then listen, they go back and forth between being the speaker and the audience, fulfilling both tasks.
Chapter 15 of the book, The Bright Side of Reason, is the chapter in which he brings everything together. As he says:
What is the context though? Immediately after this line, he goes straight into how group discussion drastically improved results of student performance on a logic problem. He then argues that it was not just the confident members of groups who swayed others, but those who provided the best arguments. He completes that subsection with this to say:
This chapter, he makes it clear that the context the theory predicts produces the greatest benefit is direct deliberation.
Under the heading “Argumentation Works”, he concludes:
LikeLike
In terms of whether it is democratic, stay tuned. My next paper will be focused on my own jury model, in isolation, and defending it’s legitimacy on it’s own terms. The paper after that is how the system relates to the demos as a whole, and it’s democratic legitimacy as a whole. There will be a wait, but I’d rather not offer these arguments piecemeal.
LikeLike
Mercier:> It is now time to make good on our promise and to show that in the right interactive context, reason works. It allows people to change each other’s minds so they end up endorsing better beliefs and making better decisions
Note that they are changing other people’s minds, not their own.
>: When argumentation lacks such demonstrative force, other factors become more relevant in convincing people or in evaluating claims, such as who seems more competent, or how many people support a given opinion.
That’s the problem (from a democratic perspective), because the decision is based on the “competence” of certain individuals and peer pressure, rather than the evaluation of the arguments. This is obviously problematic as other samples of the same population will contain different “competent” persons and information cascades will mean that numerical support could fluctuate. This is the reason that Athenian juries were charged to remain silent before registering their vote. Democracy is concerned with ensuring that the demos has kratos, not making “better” decisions.
LikeLike
He actually says “each other’s” minds. People attempt to persuade others, while they are simultaneously pursaided. This is the underlying dynamic of group discussions, and what he says it the context in which rationality works best.
Possibly. However, I propose that the right design can account for this and maintain the right amount of stability.
I believe that most people would like both. I believe both to be possible.
LikeLike
Jordan,
> confirmation bias
The gap between “prefering sources that match one’s existing ideas” to “unqualified to judge one’s own ideas” is enormous.
As I already pointed out, if we accept that people are unable to judge their own ideas then much of how we go about our lives does not make sense. It would imply that we should let others dictate their judgment to ourselves. It would also imply that the more we think about something, the less confident we should be about our own conclusions. Expertise would be no more than dogma, since the experts have spent so much time developing their ideas that their judgment on their topics of expertise are utterly unreliable.
In particular, there would be no point in having long exchanges with other on any topic since the longer the exchange proceeds, the less reliable the inferences of all those involved become.
LikeLike
Is it though? If someone only looks for evidence to support the merit of their idea, and does not look for evidence that demonstrates it’s failures, or the merits of *other* ideas, what makes them qualified to judge it?
Why do you believe other people are *unqualified* to judge it?
If a person thinks about something without engaging with others, it is very likely that they should be less confident in their ideas over time. Experts do not operate in this way. Experts are people who must convey and defend their ideas to others, and who’s answers are scrutinized. If their answers don’t survive scrutiny, they aren’t able to be recognized as experts. However, even within that domain, experts still do engage in confirmation bias, which is why the peer-reviewed system exists. Having others review the work is how we attempt to overcome confirmation bias.
This, again, does not necessarily follow. If the discussions are between like minded individuals, this is actually true, and the outcome is polarization and increasingly less reliable results. However, when there is a diversity of views, confirmation bias works in favour of this process, as each person continually brings in evidence for their own side, exposing others to that evidence and having them address it.
LikeLike
Jordan,
> Experts do not operate in this way.
Heh. Just a few comments ago you earnestly promised:
> There’s no group of “elite” who know the means better than anyone else, so I’m not sure what you are getting at.
Now, unsurprisingly, it turns out that there is an “elite” group who does know better than everyone else – experts. Their various virtues not only make them immune to the cognitive limitations of normal people, they are actually nourished by what destroys the hoi polloi. For normal people, a few months of discussions makes them unqualified to evaluate their ideas. For “experts” a lifetime of discussions only works in their favor.
LikeLike
Yoram,
Most people are experts of something, it’s simply a matter of which domain we are discussing. It would be odd, for example, for me to suggest that I understand farming better than a farmer.
However, farmers do not become experts at farming in isolation, but through being provided information by others, coming up with their own ideas and running them by others etc.
But even farmers will have various confirmation biases in regards to farming. They may believe a certain technique they came up with improves results, and they may spend great time thinking up reasons for why this technique would be of great consequence. It isn’t until they speak to another farmer to give them feedback can their argument be fully evaluated.
People, experts or otherwise, have various biases in the way they experience the world and explain it. We test those theories by talking to others to determine how strong our reasons are. If we are expecting to have to justify those reason to people who may disagree, we are more likely to antisipate various objections that may be rasied so we can determine wyas in which to counter them.
But people who have differing views, when discussing, end up producing better views on the whole. This is true for everyone. The more diverse the set of views, the stronger the arguments and reasons that will be produced.
If a group of people who have similar beliefs as one another decide to create legislation, is it entirely possible the lack of diversity will end up producing weak arguments that they were unable to see for themselves.
LikeLike
Jordan,
So, to summarize, all people without exception are unable to evaluate their own ideas. Except for experts. But we are all experts.
Discussion in a diverse group produces better ideas. But the participants in a discussion in an allotted group, which reflects the full diversity of society, is biased. It therefore needs to be evaluated by a different group, which has less opportunity for discussion. But this is an advantage because discussion produces an inability to evaluate the ideas discussed.
What a useful reasoned debate.
LikeLike
The discussion between Yoram and Jordan demonstrates that deliberative democracy doesn’t work when participants are already committed to a pre-deliberative perspective. In ATR terminology, both suffer from confirmation bias, so the exchange of reasons will have no effect on their own beliefs and preferences. If so, then the outcome of the rhetorical joust has to be determined by an agnostic audience — i.e. a large, silent jury, selected by lot.
LikeLike
Yoram, do you believe this to be an accurate description of what has been said to you?
LikeLike
Jordan, do you believe what you have written to be anything other than a voluble and obfuscated version of my summary?
LikeLike
Yes, I believe I have. So, clearly there exists some miscommunication between us.
Let’s take this sentence by sentence.
This would be an inaccurate representation of what I’ve said. Something more accurate would be “all people, without exception, have a bias in the way in which they evaluate their own ideas”.
Do you think that is a fair correction? Do you believe the distinction there is meaningful?
LikeLike
Not really. “Evaluating with a bias” results in an invalid evaluation, i.e., not an evaluation at all. In any case, the message is essentially the same.
But if you insist, let’s add a single word to my summary:
Do you now see my summary as fair?
LikeLike
Not quite. I disagree with the word “unable”. It is *possible* to evaluate your own argument, it just requires overcoming the bias. There are ways to achieve this. However, we are frequently blind to the existince of that bias, so it makes it *challenging*, but not *impossible*. Typically, when we expect that we will have to explain our arguments to others, and we expect are going to be evaluating our arguments, we are able to review our own ideas and antisipate objections people may give. In doing so, sometimes, we are actually able to find the flaws that others would find much more easily.
“Unable” is entirely the wrong word. People aren’t incapable of doing so, it’s simply *challenging* to do so.
There exists a bias when people evaluate their own ideas. A bias that is possible to overcome, but highly challenging. And also, we know people are less likely to overcome the bias when they do not antisipate needing to justify their views.
LikeLike
As a sociologist (rather than an epistemologist) I would say that justifying one’s views to one’s peers is in order to gain social approval. And the word “bias” has no meaning in a political context, as there are no right answers, only beliefs and preferences.
LikeLike
Keith, are you suggesting that confirmation bias is not a bias in a political context? How should we talk about confirmation bias, then?
LikeLike
> I disagree with the word “unable”. It is *possible* to evaluate your own argument, it just requires overcoming the bias. There are ways to achieve this. However, we are frequently blind to the existince of that bias, so it makes it *challenging*, but not *impossible*.
Careful, your elitism is showing again. The clear implication is that most people will not overcome the bias, but a select few would.
But more generally, your methodology of is fudamentally flawed. Insisting on the importance of an infinite series of nuances, sub-nuances and sub-sub-nuances is a major part of the obfuscatory style that passes for “reasoned debate”. If every word can be nuanced by adding a few dozen hundreds or thousands more words, each of which is subject to the same treatment, then the outcome of every debate is an avalanche of words that do not allow reaching any concrete or unambiguous conclusions. Thus the result of the “reasoned debate” is just more and more “reasoned debate”.
In practice, this means a preservation of the status quo. But even ignoring the conservative effect of this style of exchange, it has the very serious disadvantage of being boring. Nothing is ever learned. “Reasoned debate” arguments are mere rationalization. The “reasoned debate” technique allows one to come up with justifications to whatever conclusions one wants to come up with, or whatever conclusions one is attached to due to whatever background motivation. If one’s aim is to understand things, this activity is completely useless.
LikeLike
Everyone overcomes the bias sometimes, and doesn’t other times. There are contexts that promote overcoming it, and contexts that don’t.
Yoram, you have changed my words from one meaning to an entirely different meaning, one of which I agree with, one of which I do not. Is your expectation that I should simply lie and say I agree with the new phrasing?
LikeLike
Jordan:> Keith, are you suggesting that confirmation bias is not a bias in a political context? How should we talk about confirmation bias, then?
All I’m saying is that socialists are biased towards socialism, and conservatives conservatism (and they are unlikely to budge, however strong the arguments are). Judgment as to which is right is down the jury and the answer will be highly contingent.
LikeLike
Keith: I don’t entirely agree with that, but I think it’s a fair enough position, and I understand what you are saying.
LikeLike
Jordan,
> Is your expectation that I should simply lie and say I agree with the new phrasing?
No.
My expectation is that you make your meaning clear and exact enough so that you do not have to constantly add clarifications upon clarifications and nuances upon nuances in order to differentiate between what you mean and what are supposedly completely opposite meanings despite the fact that they are expressed with essentially the same wording.
To take the “bias” example: In the beginning everything was simple. Everybody has confirmation bias all the time, and that is that. Then, when I pointed out the absurd implications of this blanket statement, you started adding various clarifications, qualifications and nuances. Various things about experts, diversity and contexts. With these added complications, you now claim that the absurd implications do not follow from the much more convoluted and complicated new statement of the confirmation bias principle, but somehow your original conclusions (that the allotted are unqualified to evaluate their own proposals) still stand. Your new statement of the confirmation bias principle (to the extent that such a statement even exists at this point) is so complicated and vague that there is essentially no way to understand what it implies and what it does not, so one simply has to take your word for it. This is empty rhetoric and the opposite of science.
LikeLike
Yoram:> Everybody has confirmation bias all the time, and that is that.
Not so. The ATR hypothesises that confirmation bias is a valuable asset for persuaders (advocates, politicians, nerds, anoraks, obsessives and people like you and me). Most people don’t have strong views about most things, that’s why the ATR assigns them the evaluator role. The trouble with deliberative democracy is that it conflates persuading and evaluating into a single function — the pursuit of truth via communicative reasoning.
LikeLike
Yoram:
The absurd implications of the claim weren’t inherent to the claim. The clarifications made were only to explain how the implications can be otherwise.
The fact that everyone has bias, all the time, hasn’t changed. You presumed this meant people cannot overcome it and used that to dispel the notion of experts. I pointed out that’s not the implication, and explained how experts work while keeping to the original claim.
However, your summary of what we discussed did not capture the actual claims made accurately. If you felt they weren’t clearly communicated to you, that’s fair enough, but that doesn’t mean that nuance itself is the problem.
Unfortunately, nothing can be stated so perfectly and cleanly that someone attempting to find a way to misinterpret it cannot succeed. When that happens, clarifications need to be provided to “plug up the hole”, so to speak. That’s just the nature of discourse.
LikeLike
Jordan,
> The fact that everyone has bias, all the time, hasn’t changed
But, it turns out, that fact that “everyone has bias, all the time” doesn’t mean it always manifesets itself, because its manifestation depends on an ever expanding and shifting set of qualifications and conditions. To me this seems very much the same as saying not everyone has the bias all the time. But maybe not, because of some nuance or other. I can’t tell anymore. But you can rehash your endlessly-nuanced claims and arguments over and over and that is what matters, it appears.
Again, the outcome is an endless stream of words which serve as rationalizations for prejudices. This is useless as a way to learn or understand things. For some people it is useful as a way to build careers and reputations as scholars. But it is terminally degrading inellectually. Continue at your own risk.
LikeLike
Jordan >: The fact that everyone has bias, all the time
Bias from what, a transcendental Platonic standard? Confirmation bias in the ATR is simply a valuable cognitive tool to augment persuasive rhetoric. This has no implication as to whether the argument is correct or biased. And in democratic politics there is no bias, only beliefs and preferences.
LikeLike